This application proposes a two-part approach to understanding developmental aspects of the RDoC positive and negative valence systems. The project focuses on objective behavioral assessment and multimodal neuroimaging. Study 1 takes advantage of 1404 children (702 twin pairs) previously objectively assessed as 7 year-olds on positive and negative affect with a battery of videotaped behavioral measures that map onto RDoC dimensions. These children had also been assessed as toddlers and were later assessed at mean age 14 years, which allows for longitudinal tests of construct validity of the RDoC systems as applied to children. Study 1 includes completion of the age 14 data collection. Study 2 proposes to follow-up, at mean age 18 years, 640 of these participants (320 same-sex twin pairs) with comprehensive neuroimaging and concurrent psychophysiological and neuropsychological measures that assess selected RDoC constructs from the positive and negative valence systems. We will determine whether standing on the childhood RDoC measures (from Study 1) predicts adolescent neuroimaging measures of reactivity and recovery of positive and negative affect (from Study 2). We will test predictions concerning correlates of standing on RDoC constructs using structural and functional connectivity and dynamic features (reactivity and recovery) of responses to affective stimuli in an automatic emotion regulation task. We use the twin design to discern whether individual differences are genetically conditioned and whether the covariance across time (stability of RDoC) and the covariance between behavioral and neural measures (neural underpinnings of RDoC) have genetic bases. We investigate environmental bases of RDoC constructs by predicting monozygotic intrapair differences in neuroimaging parameters from earlier intrapair behavioral RDoC differences and from earlier measures of adversity and stress.